{
  "url": "https://www.aibattle.space/rankings/eval-briefcase-aa-briefcase-tool-calls-breakdown-avg-per-task",
  "dataset": "AA-Briefcase Tool Calls Breakdown, Avg per Task",
  "family": "tab::eval-briefcase",
  "familyLabel": "eval briefcase",
  "sourceRowCount": 20,
  "window": 20,
  "metricKeys": [
    "compute",
    "explore",
    "other",
    "read",
    "viewImage",
    "write"
  ],
  "rebuilt": "2026-09-19",
  "stats": {
    "median": 51.62637362637363,
    "mean": 59.319230769230764,
    "min": 31.21978021978022,
    "max": 137.16483516483515,
    "spread": 105.94505494505493,
    "gapFirstToSecond": 56.153846153846146,
    "gapFirstToSecondPercent": 69.32,
    "creators": 10
  },
  "entries": [
    {
      "rank": 1,
      "name": "Qwen3.8-Flash-Next",
      "creator": "Alibaba",
      "slug": "qwen3-8-flash-next",
      "detailsUrl": "/models/qwen3-8-flash-next",
      "value": 137.16483516483515
    },
    {
      "rank": 2,
      "name": "GLM-5.3-Flash",
      "creator": "Z AI",
      "slug": "glm-5-3-flash",
      "detailsUrl": "/models/glm-5-3-flash",
      "value": 81.01098901098901
    },
    {
      "rank": 3,
      "name": "GLM-5.3 (max)",
      "creator": "Z AI",
      "slug": "glm-5-3",
      "detailsUrl": "/models/glm-5-3",
      "value": 80.14285714285714
    },
    {
      "rank": 4,
      "name": "DeepSeek V4.1 Flash (Reasoning, Max Effort)",
      "creator": "DeepSeek",
      "slug": "deepseek-v4-1-flash",
      "detailsUrl": "/models/deepseek-v4-1-flash",
      "value": 73.57142857142857
    },
    {
      "rank": 5,
      "name": "Qwen3.8 2.4T A95B",
      "creator": "Alibaba",
      "slug": "qwen3-8-2-4t-a95b",
      "detailsUrl": "/models/qwen3-8-2-4t-a95b",
      "value": 80.41758241758242
    },
    {
      "rank": 6,
      "name": "Claude Opus 5 (Adaptive Reasoning, Max Effort)",
      "creator": "Anthropic",
      "slug": "claude-opus-5",
      "detailsUrl": "/models/claude-opus-5",
      "value": 58.78021978021978
    },
    {
      "rank": 7,
      "name": "Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)",
      "creator": "Anthropic",
      "slug": "claude-fable-5-1",
      "detailsUrl": "/models/claude-fable-5-1",
      "value": 58.26373626373626
    },
    {
      "rank": 8,
      "name": "K2 Horizon 375B A23B",
      "creator": "MBZUAI Institute of Foundation Models",
      "slug": "k2-horizon-375b-a23b",
      "detailsUrl": "/models/k2-horizon-375b-a23b",
      "value": 63.956043956043956
    },
    {
      "rank": 9,
      "name": "Kimi K3 (max)",
      "creator": "Kimi",
      "slug": "kimi-k3",
      "detailsUrl": "/models/kimi-k3",
      "value": 54.42857142857143
    },
    {
      "rank": 10,
      "name": "Qwen3.8 27B (xhigh)",
      "creator": "Alibaba",
      "slug": "qwen3-8-27b",
      "detailsUrl": "/models/qwen3-8-27b",
      "value": 52.05494505494506
    },
    {
      "rank": 11,
      "name": "Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)",
      "creator": "Anthropic",
      "slug": "claude-opus-5-xhigh",
      "detailsUrl": "/models/claude-opus-5-xhigh",
      "value": 49.75824175824176
    },
    {
      "rank": 12,
      "name": "Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)",
      "creator": "Anthropic",
      "slug": "claude-fable-5-1-xhigh",
      "detailsUrl": "/models/claude-fable-5-1-xhigh",
      "value": 50.043956043956044
    },
    {
      "rank": 13,
      "name": "MiniMax-M3",
      "creator": "MiniMax",
      "slug": "minimax-m3",
      "detailsUrl": "/models/minimax-m3",
      "value": 51.1978021978022
    },
    {
      "rank": 14,
      "name": "Muse Spark 1.3 (max)",
      "creator": "Meta",
      "slug": "muse-spark-1-3",
      "detailsUrl": "/models/muse-spark-1-3",
      "value": 42.51648351648352
    },
    {
      "rank": 15,
      "name": "GPT-5.6 Luna (max)",
      "creator": "OpenAI",
      "slug": "gpt-5-6-luna",
      "detailsUrl": "/models/gpt-5-6-luna",
      "value": 47.42857142857143
    },
    {
      "rank": 16,
      "name": "Claude Opus 5 (Adaptive Reasoning, High Effort)",
      "creator": "Anthropic",
      "slug": "claude-opus-5-high",
      "detailsUrl": "/models/claude-opus-5-high",
      "value": 41.76923076923077
    },
    {
      "rank": 17,
      "name": "Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)",
      "creator": "Anthropic",
      "slug": "claude-fable-5",
      "detailsUrl": "/models/claude-fable-5",
      "value": 40.94505494505494
    },
    {
      "rank": 18,
      "name": "DeepSeek V4 Pro 0813 (Reasoning, Max Effort)",
      "creator": "DeepSeek",
      "slug": "deepseek-v4-pro",
      "detailsUrl": "/models/deepseek-v4-pro",
      "value": 51.08791208791209
    },
    {
      "rank": 19,
      "name": "Gemini 3.8 Flash (high)",
      "creator": "Google",
      "slug": "gemini-3-8-flash",
      "detailsUrl": "/models/gemini-3-8-flash",
      "value": 31.21978021978022
    },
    {
      "rank": 20,
      "name": "GPT-5.6 Sol (max)",
      "creator": "OpenAI",
      "slug": "gpt-5-6-sol",
      "detailsUrl": "/models/gpt-5-6-sol",
      "value": 40.62637362637363
    }
  ]
}